A process for producing an insulin preparation
The present invention relates to a process for preparing a stable, quick-acting insulin preparation of the type defined in the introductory portion of claim 1. The present insulin preparation is suitable for use in insulin delivery devices.
It is generally held that most of the complications that may arise in connection with an insulin-dependent diabetes can be ascribed to inadequate control of the glucose content in the blood (Tehobroutsky, Diabetologia, 15, 143-152 (1978)). In the conventional insulin therapy, insulin is administered to the diabetic by way of one to three injections per day, resulting in a fluctuating insulin and glucose content in the blood. In contrast, the non-diabetic constantly receives insulin from his pancreas, secreted in the blood stream and in an amount adapted to his needs.
In recent years efforts have therefore been made to develop insulin delivery devices which can remedy the above-mentioned problem. If, however, such devices are applied internally or externally to a human, the insulin preparation experiences far inferior storage conditions in terms of temperature and motion when stored in the reservoir of the device than the injectable preparations.
The previously known insulin preparations are intended for storage at rest at 4ºC. In contrast, the insulin preparation is stored in the reservoir of insulin devices for an extended period of time at temperatures between 30 and 37ºC, and is moreover subjected to a good deal of motion during this period.
Thus, it is understandable that it has turned out that
in the use of the conventional dissolved insulin preparations in delivery devices, the supply tubes, valves and filters of the system are often gradually blocked by precipitated insulin aggregates, and that the desired very precise dosing of the insulin into the blood stream is prevented by this. These problems of aggregation have been discussed in great detail by W.D. Lougheed et al., Diabetologia 19, 1-9 (1980), reporting that the varying tendency to aggregation may be attributable to temperature, ion concentration and type, pH value, presence of carbon dioxide and other gases, as well as other factors.
It. has been found that the quality of the insulin reaching the blood stream after an extended stay in a delivery device with associated reservoir is significantly inferior to the quality after storage and use under the common conditions. This, too, is understandable because of the aggregation and/or formation of denatured insulin which takes place during the storage in the reservoir of the delivery device.
In the past it was attempted to solve these problems by adding to the known insulin solutions special substances - e.g. surfactants of a chain-like basic structure whose links contain alternating slightly hydrophobic and slightly hydrophilic areas, cf. the Danish Application 1851/80 - with the sole task of stabilizing the solution. Addition of such surfactants is reported to provide preparations which has such stabilizing properties as make them satisfy the technological requirements made of the use of insulin delivery devices. However, in medical circles among others there is a wish for avoiding the addition of such new substances to the known insulin preparations, which have been tested for decades, because of lack of experience with respect to the effect
of such substances on the human organism. Accordingly, there is a need for insulin preparations sufficiently stable for practical purposes without addition of special substances, even if this might result in a stability which is not of quite the same order as can be obtained by using the special additives.
The previously known dissolved insulin preparations are usually produced from zinc-crystallized insulin and are admixed with the necessary and known medium components, such as preservative, isotonic and buffer to maintain the pH, and the zinc content may optionally be adjusted with a zinc salt.
However, the art also comprises insulin preparations which are not produced on the basis of zinc-crystallized insulin. Thus, the Danish Patent Specifications 127 165, 143 024 and 143 106 teach processes, for preparing and purifying alkali metal and ammonium insulin, i.e. insulin crystallized with lithium, sodium, potassium or ammonium instead of zinc, while the Danish Patent Specification 140 801 concerns a process for preparing insulin preparations with protracted action, in which the crystallization step is omitted because immediately after the purification of crude insulin the purified insulin is passed into the preparation medium comprising an amino group containing a base, and finally the pH and optionally the zinc content are adjusted. These methods have been developed to provide better yields and preparations of lower antigenicity, respectively.
It has now surprisingly been found that an extremely stable insulin preparation, which retains its stability during shaking tests at 30 ºCfor several weeks, can be obtained by proceeding as stated in the characterizing portion of claim 1.
The invention is thus based on a combination of critical parameters:
The insulin starting material used for making the preparation must be essentially zinc-free, preferably not zinc crystallized. Useful insulins are stated in claim 2 and the preferred insulin in claim 3.
It is important that the aqueous medium does not contain di- or polyvalent ions which would impair the stability, but only monovalent ions.
The preparation medium is to be admixed with a zinc salt so that the zinc content of the finished preparation constitutes 0.5 to 1.0%, based on the weight of insulin, preferably 0.7 to 0.9%, such as 0.76 to 0.84, the optimum content being around 0.8%.
The pH of the medium is to be in the range of 6.9 to 8.9, preferably 6.9 to 7.8. The optimum range is 7.2 to 7.8, such as 7.4.
A non-dissociable isotonic, such as glycerol, or a nonreducible sugar type may be added. A specific example of a suitable sugar type is mannitol.
The preservative used may be a phenol or an alkyl- substituted phenol, preferably m-cresol.
It has been found that a preparation thus produced usually requires no addition of a buffer, in particular in case of insulin concentrations of 100 to 1500 IU/ml, because the insulin molecule itself exerts an adequate buffer effect.
However, nothing prevents the addition of common buffers, such as sodium acetate, or more special buffers of the amine or amide type, such as tyrosine amide.
Without being tied down to any specific theory, it is believed that certain problems of denaturation and aggregation observed in connection with zinc crystallization can be avoided by using an essentially zinc-free, preferably not zinc crystallized insulin as the starting material.
The subsequent addition of relatively large amounts of z.inc to the medium presumably causes formation of dissolved insulin hexamer in reversible equilibrium with monomeric and dimeric insulin.
The present process is useful in connection with insulin of different purity, such as repeatedly recrystallized products or the so-called highly purified types which are almost completely freed of antigenic impurities.
The process of the invention will be illustrated in greater detail by the following examples.
EXAMPLE 1
Sodium-crystallized insulin corresponding to 100,000 IU was dissolved in 800 ml of sterile, distilled water. Then 16 g of glycerol and 5 g of phenol were added with stirring. This was followed by addition of a solution of ZnCl2 containing an amount of zinc corresponding to 0.8% of the insulin amount, determined by Kjeldahl's method, and the solution was topped with sterile, distilled water to 990 ml. Finally, the pH of the solution was adjusted to 7.4 with 0.1 N NaOH and its volume was adjusted to 1000 ml with sterile, distilled water,
followed by sterile filtration of the solution.
EXAMPLE 2
Sodium crystallized insulin corresponding to 100,000 IU was dissolved in 200 ml of sterile distilled water, and pH was adjusted to 8.1 with 0.1 N NaOH or 0.1 N HCl. 16 g of glycerol and 3.3 g of m-cresol were dissolved in about 700 ml of sterile distilled water. This was followed by addition of a solution of ZnCl2 containing an amount of zinc corresponding to 0.8% of the insulin amount, determined by Kjeldahl's method. Then the pH of the medium was adjusted to 6.9 with 0.1 N NaOH and its volume to 790 ml with sterile distilled water. The medium was carefully added to the 200 ml of insulin solution with stirring. Finally, the pH of the solution was adjusted to 7.4 with 0.1 N NaOH, optionally 0.1 N HCl, and its volume to 1000 ml with sterile distilled water, and the solution was sterile filtrated.
EXAMPLE 3
Freeze-dried, essentially zinc-free insulin corresponding to 100,000 IU was dissolved in 800 ml of sterile, distilled water. The pH was adjusted to 7.8 with 0.1 N NaOH. The solution, was admixed, with stirring, with 1 g of NaCl and 3.0 g of m-cresol. This was followed by addition of a solution of ZnCl2 containing an amount of zinc corresponding to 0.8% of the insulin amount, determined by Kjeldahl's method, and the solution was topped with sterile, distilled water to 990 ml. Finally, the pH of the solution was adjusted to 7.8 with 0.1 N NaOH and its volume was adjusted to 1000 ml with sterile, distilled water, followed by sterile filtration of the solution.
EXAMPLE 4
A column of a diameter of 2.5 cm was packed at room temperature with a 75 cm high layer of Sephadex ® G25 medium swollen in distilled water. Of an insulin solution resulting from ion exchange in a known manner on DEAE cellulose in a 7 molar urea solution there were applied 40 ml containing insulin corresponding to 20,000 IU after concentration and adjustment of the pH of the solution to 8.2. This was followed by elution with distilled water at a rate of 108 ml/h, and the eluate was divided into fractions. The extinction at 280 run was measured, and the fractions containing insulin were collected, followed by determination of the insulin content. Part of the resulting insulin solution containing insulin corresponding to 10,000 IU was diluted with sterile, distilled water to 80 ml. The solution was admixed, with stirring, with 1.6 g of glycerol, 0.2 g of NaCl and 0.33 g of m-cresol. Then there was added a solution of ZnCl2 containing an amount of zinc corresponding to 0.8% of the insulin amount, determined by Kjeldahl's method, and the solution was topped with sterile, distilled water to 95 ml. Finally, the pH of the solution was adjusted, to 7.4 with 0.1 N NaOH, and its volume was adjusted -to 100 ml with sterile, distilled water, followed by sterile filtration of the solution.
The preparations described in the examples were tested for stability by shaking tests at 30 ºCwith 70 os/min in horizontal 3/4 filled vials, which is tantamount to extreme conditions compared to the storage in the reservoir in a delivery device.
The preparations produced in examples 1, 3 and 4 were found to be stable for more than 360 hours, while example
2 provides for stability of about 1500 hours.
Common commercial quick-acting neutral insulin solutions were found to have inferior stability in corresponding tests. Thus, the stability of an insulin preparation produced as described in example 1 of the US Patent Specification 3 091 573 lasts for about 24 hours. The stability of an insulin preparation produced according to example 1 of the Danish Patent Specification 95007 was determined to be about 97 hours.